The Romanian Alphabet

The Romanian alphabet is the Latin alphabet you already know, with five extra letters carrying diacritics: ă, â, î, ș, ț. The single most important thing to understand on day one is that these five are full, separate letters — not decorated versions of a, i, s, and t, and not accents you can drop when you're in a hurry. They have their own slots in the alphabet, their own places in the dictionary, and their own sounds. Treat Romanian as "Italian with a few squiggles" and you will misspell, mispronounce, and mis-sort words from the start. Treat it as a 31-letter alphabet and everything lines up.

The 31 letters

Counting the five diacritic letters, the Romanian alphabet has 31 letters. Here they are in official alphabetical order, which is exactly where Romanian dictionaries and indexes put them — note that each diacritic letter sits immediately after its base letter, not at the end of the alphabet:

The five special letters are letters, not accents

This is the conceptual hurdle for English (and Italian, French, Spanish) speakers. In French, the accent on é is a modification of an existing letter e; alphabetize a French word list and é sorts with e. Romanian works differently: ă and â are distinct letters from a, with their own dictionary positions and — crucially — their own sounds that a does not make.

• ă /ə/ is the "schwa," the unstressed vowel in English "sofa" — but in Romanian it can also be stressed. casă ("house") ends in this sound.
• â /ɨ/ is a high central vowel with no English equivalent (see Why diacritics matter). român ("Romanian") has it in the middle.
• î /ɨ/ is the same sound as â — the two letters split the work positionally, covered fully on the î vs â rule page.
• ș /ʃ/ is English "sh." ușă ("door").
• ț /ts/ is the "ts" in "cats." preț ("price").

Because each is a separate sound, swapping the diacritic letter for its base letter produces a different word, not a misspelling of the same word. casa (with plain a) means "the house"; casă (with ă) means "a house / house." They are not interchangeable.

k, q, w, y: the loanword corner

Four Latin letters — k, q, w, y — exist in the alphabet but are almost never used in native Romanian words. Native Romanian writes the /k/ sound with c (or ch before e/i), so k shows up only in borrowings and names: kilogram, kaki, karate, whisky, weekend, yală ("Yale lock"), quasar. This is the opposite of English, where k, w, and y are core letters.

No digraph letters — but ch and gh are spelling devices

Unlike Spanish (which long treated ch and ll as letters) or Hungarian, Romanian has no digraphs that count as letters. Every letter in the table above is a single character. But Romanian does use two letter combinations as a pure spelling trick, and you must recognize them.

Because c and g go soft before e and i (ce = "cheh," gi = "ji"), Romanian needs a way to write the hard /k/ and /ɡ/ sounds before those same vowels. The solution is to insert a silent h: ch = hard /k/, gh = hard /ɡ/, but only before e and i.

• che, chi = /ke, ki/ — chitară ("guitar") = "kee-TA-ră," ureche ("ear") = "u-RE-ke."
• ghe, ghi = /ɡe, ɡi/ — ghid ("guide") = "geed," unghie ("fingernail") = "UN-gee-e."

The h here is not a sound — it is a shield that blocks the softening. This is the one place where Romanian spelling looks like it has a digraph, but ch and gh are not letters; they are c and g wearing a hard hat. See the c and g soft/hard rule for the full mechanics.

Why the alphabet was built this way

Romanian only adopted the Latin alphabet officially in the 1860s, replacing the Cyrillic script it had used for centuries. The designers built the new orthography to be phonemic — close to one letter, one sound — and to make the language's Latin ancestry visible. That is why the five diacritic letters exist: Romanian has sounds (the central vowels /ə/ and /ɨ/, and /ʃ/, /ts/) that plain Latin letters could not represent, so new letters were created rather than borrowing inconsistent digraphs the way English does (sh, ch, th). The loanword letters k, q, w, y were kept in the alphabet for foreign words but pointedly left out of the native system. The whole design is deliberate: a tight, mostly one-to-one mapping between letters and sounds, with the five diacritics doing the heavy lifting.

Common Mistakes

Treating ă/â/î/ș/ț as optional "accented" versions of a/i/s/t:

Writing the /k/ sound with k in native words (English/German habit):

Alphabetizing the diacritic letters at the end (or merging them with the base):

Reading ch/gh as the English "church" / soft-g sounds:

Key Takeaways

• The Romanian alphabet has 31 letters: the 26 Latin letters plus ă, â, î, ș, ț.
• The five diacritic letters are full letters with their own sounds and dictionary positions — they sort immediately after their base letter, not at the end.
• k, q, w, y exist but are used only in loanwords; native /k/ is written c (or ch), native /v/ is v, native /i/ is i.
• Romanian has no digraph letters, but ch = hard /k/ and gh = hard /ɡ/ before e/i — a spelling device, not new letters.
• The orthography was built in the 1860s to be phonemic and to display the language's Latin roots, which is exactly why the special letters were invented.